BMC Public Health · Nov 18, 2026
Lifestyle risk behaviours and multimorbidity in older adults: evidence from WHO SAGE India Wave 3
By Chinnaiyan S, Palanisamy B
Background Although lifestyle risk behaviours are recognised determinants of chronic disease, evidence on their combined association with multimorbidity among older adults in India remains limited. This study examined the prevalence of lifestyle risk behaviours and multimorbidity and their combined association among older adults in India. Methods This cross-sectional study used survey-weighted WHO SAGE India Wave 3 data from 6,719 adults aged ≥ 50 years across six states. Multimorbidity was defined as ≥ 2 diagnosed chronic conditions. Five lifestyle risk behaviours were combined into low-, moderate-, and high-risk categories. Survey-weighted logistic regression assessed crude and adjusted associations and interactions. Results Multimorbidity prevalence was 17.7% (95% CI 15.7–19.7), with hypertension (29.0%, 95% CI 26.8–31.3) and arthritis (12.9%, 95% CI 11.4–14.4) being the most prevalent conditions. High behavioural risk was present in 16.5% (95% CI 14.8–18.2). Compared with low risk, moderate and high behavioural risk showed lower crude odds of multimorbidity (OR 0.69, 95% CI 0.47–1.01; OR 0.68, 95% CI 0.45–1.01), but associations were not statistically significant after adjustment (aOR 0.79, 95% CI 0.54–1.16 for moderate risk; aOR 0.78, 95% CI 0.51–1.20 for high risk). Age, wealth, Scheduled Tribe status, and state were independently associated with multimorbidity. Exploratory interaction analyses provided no statistically robust evidence of effect modification by age, sex, residence, or wealth quintile after Bonferroni correction. Conclusion Combined lifestyle risk behaviours were not independently associated with diagnosed multimorbidity. Sociodemographic characteristics, particularly age, wealth, and state, were independently associated with multimorbidity. Longitudinal studies with biomarker-confirmed disease ascertainment are needed to clarify these relationships. These findings highlight the need to interpret cross-sectional behavioural associations cautiously, particularly when chronic conditions are based on prior diagnosis and healthcare contact.